Matlab-based Framework for Traffic Modelling Using Hybrid Petri Nets

OData support
Supervisor:
Kovács Gábor
Department of Control Engineering and Information Technology

The increasing number of vehicles has made congestions more and more frequent, even on high-capacity motorways. Avoidance of congestions or minimizing their effects is of paramount importance due to economical and environmental reasons. Various methods have been proposed for congestion control, which need detailed models of the road networks. One of the modeling approaches found in the literature is the use of hybrid Petri nets.

This BSc Thesis presents a MATLAB-based framework for modeling motorway segments and traffic flow using hybrid Petri nets. In order to fit actual traffic data, collected on the physical road network, the framework provides wide parameterization of segments. Modeling and simulation is supported by an easy-to-use graphical user interface and by the graphical representation of simulation results.

Chapter 2 summarizes the most important preliminaries about Petri-nets, especially timed hybrid Petri-nets. Modeling and simulation of Petri-nets in MATLAB environment as well as their extensions to meet the needs of traffic modeling are presented by Chapter 3.

Basic components of motorway modeling are given by Chapter 4. Besides the components themselves, their modeling by Petri-nets is also presented in details. The MATLAB functions developed for the use of these models in traffic modeling are summarized in Chapter 5.

Chapter 6 presents the graphical user interface, which makes the construction of the motorway model easy and straightforward. Besides supporting the construction of the model, the graphical user interface is also capable of generating a MATLAB code representing the modeled motorway segment.

Chapter 7 presents the capabilities of the toolbox through two examples. Details of modeling and simulation are given also in graphical form. Future work is summarized by Chapter 8.

Downloads

Please sign in to download the files of this thesis.